|
I get "SR171: Transaction timed out" on the dbpedia sparql endpoint, for the following simple query: Why do I get this error? I don't set any timeout - it's on 0.
|
|
Here is Your Sparql Query as URL to the DbPedia sparql endpoint. Which just seems to work. Maybe there was a temporary overload of the system that made it timeout, or maybe you have to add additional information on how you operate your query? |
|
If you really want to have fun with DBpedia, you need to load it into your own triple store. If you've got a machine with an above-average amount of memory (like my 24GB Windows workstation or a high-memory extas large instance in the AMZN cloud) it's pretty easy to load DBpedia into the open source edition of Virtuoso -- the DBpedia public endpoint uses the commercial version of the same software, so you should get very much the same results. I've loaded data sets of similar size into an 8GB laptop and had success, although the loading takes most of a day. DBpedia has some anomalies and Virtuoso is more tolerant of broken RDF than some other products, so you might have some challenges with other triple stores. I know Ontotext has a special edition of DBpedia that has been cleaned to work with OWLIM-SE. Note also that the version of DBpedia in the public endpoint is incomplete: it lacks the "Wikipedia pagelinks" data set, which I find particularly fruitful. |
|
I guess missing the trailing Your query is quite expensive, requiring to check through the birthdates of all people in Wikipedia and apply the filter you specify, hitting less that 1-in-365 persons. Try starting with a
Increment the limit or fiddle with the range (as a cheap version of "offset") until you're happy with the answers you get. |
|
Gerrit V: I run the query from the browser (FF) with the http://dbpedia.org/sparql interface and it gives the timeout error. Even a simple Querida like the one below ( i didn't include the namespaces-they are the same as above) with LIMIT 1 gives timeout. I also run the query from PHP - (wich is where i want to use it ultimately). But the query results are the same in the browser as from PHP. So it either works/doesn't work in both. Gerrit V: yes the problem is that dbpedia's system is overloaded. The query works late at night or in the morning (i'm in Romania Eastern Europe). It doesn't work in the rest of the time. This is not acceptable for the website of course. Signified: Using LIMIT 1 to limit the answers is probably a good idea. However on my system and from my hosting's system it doesn't work even with LIMIT 1 (-> one). I still get the timeout error. Wikipedia works with a lot less money than other sites so I think this is why their system is so limited. I don't think there is much I can do. I have to give up on this DBPedia related features wich is too bad, but I don't think there's much I can do. If anyone has sugestions I am happy to try them out. Thank you all for your answers. Yes the problem is that dbpedia's system is overloaded. What is interesting is that it hardly takes 1-2 seconds for the query to execute and i see the error. DB Pedia could allow at least 2-3 seconds for a query. I'm not sure why they don't. The query works late at night or in the morning (i'm in Romania Eastern Europe). It doesn't work in the rest of the time. This is not acceptable for the website of course. I have a question though: Does it work from anyone's system at usual hours? Please try and tell me. Thanks
|



Duplicated on http://stackoverflow.com/questions/9176596/simple-sparql-query-not-working-dates-comparison